This is our old blog. It hasn't been active since 2011. Please see the link above for our current blog or click the logo above to see all of the great data and content on this site.

There are really two charts

Posted by Doug on May 2, 2007

A quick analysis of this weekend's trades indicate that teams do stay pretty close to the pick value chart when exchanging picks in the same draft. For example, take a look at the second Denver/Jacksonville swap in the middle of the first round. Denver gets #17, which is worth 950 points, while Jacksonville gets #21, #86, and #198, worth a total of 972 points. Most of the other trades from day one show a similar pattern of following the chart closely. Here's a list:

High pick is the earliest pick that changed hands in the trade. Up and down represent the teams moving up and down respectively in the deal, and the pick chart value of the picks they received.

high pick      up          down
   14         NYJ 1115    CAR 1056
   17         DEN  950    JAX  972
   26         DAL  700    PHI  723
   33         ARI  580    OAK  604
   34         BUF  560    DET  690
   41         ATL  490    MIN  512
   47         NYJ  430    GB   436
   53         CLE  383    DAL  363
   58         DET  320    NOR  294
   62         DET  284    BAL  316
   86         BAL  160    JAX  131

I don't know the history of the pick chart, but I assume it was built to match the market that had already been established. In other words, it seems likely that the chart is based on trades that had actually occurred.

But the vast majority of those trades occurred while one of the two teams was on the clock, and therefore both teams know exactly who is available with the pick. My strong suspicion is that Jacksonville would have been happy to make this deal on Saturday morning, while Denver would not have. The Broncos were only willing to make it because they knew for sure that a particular player --- Jarvis Moss in this case --- was available. And this is the case for most of the draft day trades.

Since the team trading up is the one with their sites on a particular guy that they now know is available, they're the team that's more likely to overpay (compared to what they would have paid 24 hours prior). But because the chart is built from historical data, this overpayment has been built right into the chart!

Really, there ought to be two charts: an on-the-clock chart and a pre-draft chart. The one we always see is the on-the-clock chart. I think it's wrong to read it as "this is the standard rate, but I should be willing to pay more to move up if I really like the guy who's available." Rather, it reflects the most that you should be willing to pay to move up. Or at least it reflects what teams have been willing to pay when they know they really like the guy who's available.